Ion implanters are widely used in semiconductor manufacturing to selectively alter the conductivity of materials. In a typical ion implanter, ions generated from an ion source are transported downstream through a series of beamline components which may include one or more analyzer and/or corrector magnets and a plurality of electrodes. The analyzer magnets may be used to select desired ion species and filter out contaminant species or ions having undesirable energies. The corrector magnets may be used to manipulate the shape of the ion beam or otherwise adjust the ion beam quality before it reaches a target wafer. Suitably shaped electrodes can be used to modify the energy and the shape of the ion beam. After the ion beam has been transported through the series of beamline components, it may be directed into an end station to perform ion implantation.
FIG. 1 depicts a conventional ion implanter system 100. As is typical for most ion implanters, the system 100 is housed in a high-vacuum environment. The ion implanter system 100 may comprise an ion source 102 and a series of beamline components through which an ion beam 10 passes. The series of beamline components may include, for example, an extraction manipulator 104, a filter magnet 106, an acceleration or deceleration column 108, an analyzer magnet 110, a rotating mass slit 112, a scanner 114, and a corrector magnet 116. Much like a series of optical lenses that manipulate a light beam, the ion implanter components can filter and focus the ion beam 10 before steering it towards a target wafer 118. The target wafer 118 is typically housed in a wafer end-station (not shown) under high vacuum.
In semiconductor manufacturing, ion implantation of a target wafer is often performed on only selected areas of the wafer surface, while the rest of the wafer surface is typically masked with a photosensitive material known as “photoresist.” Through a photolithography process, the target wafer may be coated with a patterned layer of photoresist material, exposing only selected areas of the wafer surface where ion implantation is desired. During ion implantation, an ion beam makes its impact not only on the exposed portion of the wafer surface, but also on the photoresist layer. The energetic ions often break up chemical bonds within the photoresist material and release volatile organic chemicals and/or other particles into the vacuum chamber (i.e., wafer end-station) that houses the target wafer. This phenomenon is known as “photoresist outgassing.”
Photoresist outgassing in an ion implanter can have several deleterious effects on an ion beam. For example, the particles released from the photoresist may cause a pressure increase or pressure fluctuations in the high-vacuum wafer end-station. The outgassed particles may also migrate upstream from the wafer end-station to other beamline components, such as the corrector magnet 116 and the scanner 114 as shown in FIG. 1, and may affect vacuum levels in those portions of the ion implanter as well.
The outgassed particles often interact with an incident ion beam, for example, by exchanging charges with beam ions. As a result, an ion with a single positive charge may lose its charge to an outgassed particle and become neutralized; a doubly charged ion may lose one positive charge to an outgassed particle and become singly charged; and so on. As a result, the outgassing-induced charge exchange can interfere with the ion dosimetry system in the ion implanter. A typical ion dosimetry system determines ion doses by integrating a measured beam current over time and converting the integrated beam current (i.e., total ion charges) to a total dose based on an assumption that a particular ion species has a known charge state. The outgassing-induced charge exchange, however, randomly alters the charge state of the ion species, thereby invalidating the charge-state assumption that the ion dosimetry system relies on. For example, if the outgassed particles tend to rob positive charges from positive ions, then such charge exchange will cause the dosimetry system to undercount that ion species, which in turn leads to an over-supply of that ion species to a target wafer.
Due to the above-mentioned upstream migration of outgassed particles, some of the charge exchange may occur in a corrector magnet. Charge-altered ions are subject to a different Lorentz force as compared to those same species of ions that experience no charge exchange. As such, the charge-altered ions will deviate from the main ion beam path, resulting in non-uniform dosing of the target wafer. Beamlets formed by streams of the charge-altered ions are referred to hereinafter as “parasitic beamlets.” Apart from generating the parasitic beamlets, the charge exchange can also alter energies and angles of the affected ions, both of which will affect the ultimate dopant profile in the target wafer.
As those skilled in the art will readily appreciate, the above-described parasitic beamlet problems may similarly arise in an ion implanter running a molecular ion beam. Interactions of the molecular ion beam with residual gases in the ion implanter may cause molecular breakups, resulting in ions with altered charges as well as altered masses. Therefore, the molecular breakups can also introduce contamination to the ion implantation process.
A number of techniques have been developed to alleviate the effects of outgassing-induced charge exchange. For example, to compensate for the effect of charge exchange on dosimetry, it has been proposed that an direct current (DC) offset be included in the beam current integration process, the DC offset being correlated to a gas pressure in the wafer end-station. However, this dosimetry compensation approach only addresses one aspect of the problems caused by parasitic beamlets.
According to another approach, the above-mentioned pressure fluctuation caused by photoresist outgassing may be mitigated by bleeding inert gases into a wafer end-station in an amount much greater than the level of outgassing. While this method might stabilize the gas pressure in the wafer end-station, the resulting higher-than-optimum pressure may negatively affect the ion implantation.
According to yet another approach as illustrated in FIG. 2, a conductance-limiting aperture 202 may be provided between a wafer end-station 204 and beamline components 206. The conductance-limiting aperture 202 is typically a fixed aperture that is just wide enough to allow a scanned ion beam 20 to pass through. The conductance-limiting aperture 202 is intended to serve two purposes, i.e., to reduce an upstream migration of outgassed particles and to block parasitic beamlets from entering the wafer end-station 204. However, since the conductance-limiting aperture 202 has to be wider than the diameter of a target wafer 208, this approach can achieve only limited success in reducing conductance of outgassed particles or blocking parasitic beamlets.
In view of the foregoing, it would be desirable to provide a solution to reduce effects of photoresist outgassing which overcomes the above-described inadequacies and shortcomings.